npm - @fugood/llama.node - Versions diffs - 0.3.8 → 0.3.10 - Mend

@fugood/llama.node 0.3.8 → 0.3.10

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (105) hide show

package/bin/darwin/arm64/llama-node.node +0 -0
package/bin/darwin/x64/llama-node.node +0 -0
package/bin/linux/arm64/llama-node.node +0 -0
package/bin/linux/x64/llama-node.node +0 -0
package/bin/linux-cuda/arm64/llama-node.node +0 -0
package/bin/linux-cuda/x64/llama-node.node +0 -0
package/bin/linux-vulkan/arm64/llama-node.node +0 -0
package/bin/linux-vulkan/x64/llama-node.node +0 -0
package/bin/win32/arm64/llama-node.node +0 -0
package/bin/win32/arm64/node.lib +0 -0
package/bin/win32/x64/llama-node.node +0 -0
package/bin/win32/x64/node.lib +0 -0
package/bin/win32-vulkan/arm64/llama-node.node +0 -0
package/bin/win32-vulkan/arm64/node.lib +0 -0
package/bin/win32-vulkan/x64/llama-node.node +0 -0
package/bin/win32-vulkan/x64/node.lib +0 -0
package/lib/binding.js +2 -2
package/lib/binding.ts +52 -8
package/lib/index.ts +3 -1
package/package.json +8 -1
package/src/LlamaCompletionWorker.cpp +33 -6
package/src/LlamaCompletionWorker.h +3 -1
package/src/LlamaContext.cpp +387 -28
package/src/LlamaContext.h +5 -0
package/src/common.hpp +19 -2
package/src/llama.cpp/.github/workflows/build.yml +289 -107
package/src/llama.cpp/.github/workflows/close-issue.yml +1 -1
package/src/llama.cpp/.github/workflows/docker.yml +2 -1
package/src/llama.cpp/.github/workflows/server.yml +25 -2
package/src/llama.cpp/CMakeLists.txt +10 -19
package/src/llama.cpp/cmake/build-info.cmake +1 -1
package/src/llama.cpp/common/CMakeLists.txt +32 -0
package/src/llama.cpp/common/arg.cpp +66 -16
package/src/llama.cpp/common/chat-template.hpp +515 -0
package/src/llama.cpp/common/chat.cpp +966 -0
package/src/llama.cpp/common/chat.hpp +52 -0
package/src/llama.cpp/common/common.cpp +159 -36
package/src/llama.cpp/common/common.h +56 -14
package/src/llama.cpp/common/json-schema-to-grammar.cpp +46 -66
package/src/llama.cpp/common/json-schema-to-grammar.h +15 -1
package/src/llama.cpp/common/llguidance.cpp +270 -0
package/src/llama.cpp/common/log.cpp +1 -10
package/src/llama.cpp/common/log.h +10 -0
package/src/llama.cpp/common/minja.hpp +2868 -0
package/src/llama.cpp/common/sampling.cpp +22 -1
package/src/llama.cpp/common/sampling.h +3 -0
package/src/llama.cpp/docs/build.md +54 -9
package/src/llama.cpp/examples/export-lora/export-lora.cpp +12 -2
package/src/llama.cpp/examples/gbnf-validator/gbnf-validator.cpp +1 -1
package/src/llama.cpp/examples/llava/CMakeLists.txt +7 -0
package/src/llama.cpp/examples/llava/clip-quantize-cli.cpp +59 -0
package/src/llama.cpp/examples/llava/clip.cpp +133 -14
package/src/llama.cpp/examples/llava/clip.h +2 -0
package/src/llama.cpp/examples/llava/llava.cpp +22 -8
package/src/llama.cpp/examples/llava/minicpmv-cli.cpp +9 -1
package/src/llama.cpp/examples/main/main.cpp +26 -25
package/src/llama.cpp/examples/run/linenoise.cpp/linenoise.cpp +136 -137
package/src/llama.cpp/examples/run/linenoise.cpp/linenoise.h +18 -4
package/src/llama.cpp/examples/run/run.cpp +224 -69
package/src/llama.cpp/examples/server/server.cpp +252 -81
package/src/llama.cpp/examples/server/utils.hpp +73 -21
package/src/llama.cpp/examples/simple-chat/simple-chat.cpp +6 -4
package/src/llama.cpp/examples/simple-cmake-pkg/CMakeLists.txt +11 -0
package/src/llama.cpp/ggml/CMakeLists.txt +78 -1
package/src/llama.cpp/ggml/include/ggml.h +1 -1
package/src/llama.cpp/ggml/src/CMakeLists.txt +21 -4
package/src/llama.cpp/ggml/src/ggml-alloc.c +1 -13
package/src/llama.cpp/ggml/src/ggml-cpu/ggml-cpu-quants.c +91 -78
package/src/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c +7 -7
package/src/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.cpp +2 -1
package/src/llama.cpp/ggml/src/ggml-cuda/CMakeLists.txt +1 -1
package/src/llama.cpp/ggml/src/ggml-cuda/vendors/hip.h +46 -0
package/src/llama.cpp/ggml/src/ggml-hip/CMakeLists.txt +16 -1
package/src/llama.cpp/ggml/src/ggml-musa/CMakeLists.txt +1 -1
package/src/llama.cpp/ggml/src/ggml-rpc/ggml-rpc.cpp +28 -8
package/src/llama.cpp/ggml/src/ggml-sycl/ggml-sycl.cpp +5 -7
package/src/llama.cpp/ggml/src/ggml-sycl/softmax.cpp +33 -23
package/src/llama.cpp/ggml/src/ggml-sycl/softmax.hpp +1 -5
package/src/llama.cpp/ggml/src/ggml-vulkan/ggml-vulkan.cpp +323 -121
package/src/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp +13 -3
package/src/llama.cpp/ggml/src/ggml.c +23 -13
package/src/llama.cpp/include/llama.h +14 -1
package/src/llama.cpp/models/ggml-vocab-deepseek-r1-qwen.gguf.inp +112 -0
package/src/llama.cpp/models/ggml-vocab-deepseek-r1-qwen.gguf.out +46 -0
package/src/llama.cpp/src/CMakeLists.txt +1 -1
package/src/llama.cpp/src/llama-arch.cpp +7 -2
package/src/llama.cpp/src/llama-arch.h +3 -1
package/src/llama.cpp/src/llama-chat.cpp +11 -2
package/src/llama.cpp/src/llama-chat.h +1 -0
package/src/llama.cpp/src/llama-grammar.cpp +86 -6
package/src/llama.cpp/src/llama-grammar.h +22 -1
package/src/llama.cpp/src/llama-mmap.cpp +1 -0
package/src/llama.cpp/src/llama-model-loader.cpp +1 -1
package/src/llama.cpp/src/llama-model.cpp +76 -6
package/src/llama.cpp/src/llama-sampling.cpp +47 -4
package/src/llama.cpp/src/llama-vocab.cpp +10 -4
package/src/llama.cpp/src/llama.cpp +181 -123
package/src/llama.cpp/tests/CMakeLists.txt +4 -0
package/src/llama.cpp/tests/test-backend-ops.cpp +158 -57
package/src/llama.cpp/tests/test-chat-template.cpp +154 -31
package/src/llama.cpp/tests/test-chat.cpp +607 -0
package/src/llama.cpp/tests/test-grammar-integration.cpp +2 -2
package/src/llama.cpp/tests/test-grammar-llguidance.cpp +1140 -0
package/src/llama.cpp/tests/test-json-schema-to-grammar.cpp +1 -1
package/src/llama.cpp/examples/main-cmake-pkg/CMakeLists.txt +0 -32

package/src/llama.cpp/tests/test-json-schema-to-grammar.cpp CHANGED Viewed

@@ -1246,7 +1246,7 @@ int main() {
     test_all("C++", [](const TestCase & tc) {
         try {
-            tc.verify(json_schema_to_grammar(nlohmann::ordered_json::parse(tc.schema)));
+            tc.verify(json_schema_to_grammar(nlohmann::ordered_json::parse(tc.schema), true));
             tc.verify_status(SUCCESS);
         } catch (const std::runtime_error & ex) {
             fprintf(stderr, "Error: %s\n", ex.what());

package/src/llama.cpp/examples/main-cmake-pkg/CMakeLists.txt DELETED Viewed

@@ -1,32 +0,0 @@
-cmake_minimum_required(VERSION 3.12)
-project("llama-cli-cmake-pkg" C CXX)
-set(TARGET llama-cli-cmake-pkg)
-find_package(Llama 0.0.1 REQUIRED)
-# Bake common functionality in with target. Because applications
-# using the relocatable Llama package should be outside of the
-# source tree, llama-cli-cmake-pkg pretends the dependencies are built-in.
-set(_common_path "${CMAKE_CURRENT_LIST_DIR}/../../common")
-add_library(common OBJECT)
-file(GLOB _common_files
-    "${_common_path}/*.h"
-    "${_common_path}/*.cpp"
-)
-target_sources(common PRIVATE ${_common_files})
-# If the common project was part of "llama-cli-cmake-pkg" the transient
-# defines would automatically be attached. Because the common func-
-# tionality is separate, but dependent upon the defines, it must be
-# explicitly extracted from the "llama" target.
-#
-get_target_property(_llama_transient_defines llama
-    INTERFACE_COMPILE_DEFINITIONS)
-target_compile_definitions(common PRIVATE "${_llama_transient_defines}")
-add_executable(${TARGET} ${CMAKE_CURRENT_LIST_DIR}/../main/main.cpp)
-target_include_directories(${TARGET} PRIVATE ${_common_path})
-install(TARGETS ${TARGET} RUNTIME)
-target_link_libraries(${TARGET} PRIVATE common llama ${CMAKE_THREAD_LIBS_INIT})
-target_compile_features(${TARGET} PRIVATE cxx_std_17)